Supports integration with data preparation and feature engineering pipelines, including versioning of datasets and capabilities used in ML experiments.
Data Pipeline Integration
CCC.MLDE.CP07
Related Threats
| ID | Title | Description |
|---|---|---|
| CCC.MLDE.TH02 | Training Data or Model Artifacts are Exfiltrated | Notebook instances and connected data pipelines may be configured with unrestricted egress paths, such as file downloads, public network interfaces, or writable external destinations. Training datasets, model artifacts, and embedded credentials could be transferred out of the environment through these paths. This results in a loss of confidentiality for proprietary data and models, and may expose regulated or sensitive records used in training. |